arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.05195cs.AI
收藏

可追溯的多源RAG用于藏医药问答

Traceable Cross-Source RAG for Chinese Tibetan Medicine Question Answering

School of Information Science & Engineering, Lanzhou University(信息科学与工程学院,兰州大学)

Fengxian Chen, Zhilong Tao, Jiaxuan Li, Yunlong Li, Qingguo Zhou

AI总结:

本文提出DAKS和对齐图方法,用于提升多源RAG在藏医药问答中的可追溯性、减少幻觉并增强跨知识库验证能力。

摘要:

检索增强生成(RAG)承诺提供基于事实的问题回答,但涉及多个异构知识库(KBs)的领域设置仍具挑战性。在藏医药中,百科条目通常密集且容易匹配,这可能会主导检索,即使经典或临床论文提供更权威的证据。我们研究了一个包含三个知识库(百科、经典和临床论文)的实用设置,并有一个500个查询基准(截止K=5),涵盖单一KB和跨KB问题。我们提出了两种互补的方法来提高可追溯性,减少幻觉,并启用跨KB验证。首先,DAKS执行KB路由和预算检索,以缓解密度驱动的偏差,并在适当的情况下优先考虑权威来源。其次,我们使用对齐图来指导证据融合和覆盖感知打包,提高跨KB证据覆盖,而无需依赖于朴素的连接。所有答案均由轻量级生成器openPangu-Embedded-7B生成。实验显示在路由质量和跨KB证据覆盖方面有持续的增益,全系统在CrossEv@5上取得最佳成绩,同时保持强忠实度和引用正确性。

英文摘要:

Retrieval-augmented generation (RAG) promises grounded question answering, yet domain settings with multiple heterogeneous knowledge bases (KBs) remain challenging. In Chinese Tibetan medicine, encyclopedia entries are often dense and easy to match, which can dominate retrieval even when classics or clinical papers provide more authoritative evidence. We study a practical setting with three KBs (encyclopedia, classics, and clinical papers) and a 500-query benchmark (cutoff $K{=}5$) covering both single-KB and cross-KB questions. We propose two complementary methods to improve traceability, reduce hallucinations, and enable cross-KB verification. First, DAKS performs KB routing and budgeted retrieval to mitigate density-driven bias and to prioritize authoritative sources when appropriate. Second, we use an alignment graph to guide evidence fusion and coverage-aware packing, improving cross-KB evidence coverage without relying on naive concatenation. All answers are generated by a lightweight generator, \textsc{openPangu-Embedded-7B}. Experiments show consistent gains in routing quality and cross-KB evidence coverage, with the full system achieving the best CrossEv@5 while maintaining strong faithfulness and citation correctness.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑